Immune checkpoint therapies have led to remarkable clinical responses in human cancers such as melanoma, colorectal cancer, and non-small cell lung cancer. Key to efficacious use of these treatments, is patient selection based on molecular indicators that faithfully predict response. With a need for targeted therapy, identifying biomarkers to bring the clinical benefits of immune checkpoint therapy to triple negative breast cancer (TNBC) promises to impact patient outcomes and is the goal of this proposal. My central hypothesis is that immune checkpoint therapy can be extended to some breast cancer patients with the assistance of appropriate biomarker selection coming from pre-clinical testing. Using predictive bioinformatics analysis of 819 human breast cancer samples, I found evidence that patients with basal-like TNBC present similar clinical indicators as other cancers that benefit from anti-CTLA4 and anti-PD1 therapy. This indicates a high likelihood for efficacious treatment of TNBC patients with these therapies. By establishing the appropriate preclinical models, simulating clinical trials with immune checkpoint therapies can be accomplished. Indeed, preliminary data reveals a therapeutic benefit when combining anti-PD1 and common use chemotherapy in one of the mouse models that credentialed for basal-like gene expression features and expression of key immune signatures. The clinical and translational value in simulating clinical trials using genetically engineered mouse models (GEMMs) is the ability to test combination therapies and utilize response data to develop predictive biomarkers that can be carried over to guide future clinical trials. Based on these findings and goals, I have developed two aims. In aim 1, I will test the hypothesis that TNBC GEMMs will present variable responses to immune checkpoint inhibitors, and that by using sensitive and resistant GEMMs I will be able to identify genomic biomarkers of response that will be leveraged in comparative analyses with human cancers. My goal is to use this setting to identify biomarkers and treatment strategies that will allow these therapies to be used to treat TNBC. In aim 2, I will test the hypothesis that Apobec3 expression will lead to many new mutations and a unique DNA mutagenesis signature, and potentially many neoantigens, which will sensitize tumors to immune checkpoint inhibitors. The result of this work will determine whether the human homolog, Apobec3B, is a molecular predictor for application of immune checkpoint therapies. The work in this proposal aligns itself with the mission of the NCI as the outcomes are expected to inform future clinical trials using immune checkpoint therapies and improve survival for breast cancer patients.